Residual LSTM-based short duration forecasting of polarization current for effective assessment of transformers insulation
Abstract The empirical application of polarization and depolarization current (PDC) measurement of transformers facilitates the extraction of critical insulation-sensitive parameters. This technique, rooted in time-domain dielectric response analysis, forms the bedrock for parameterization and insul...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2024-01-01
|
| Series: | Scientific Reports |
| Online Access: | https://doi.org/10.1038/s41598-023-50641-z |
| _version_ | 1797350030574616576 |
|---|---|
| author | Aniket Vatsa Ananda Shankar Hati Prashant Kumar Martin Margala Prasun Chakrabarti |
| author_facet | Aniket Vatsa Ananda Shankar Hati Prashant Kumar Martin Margala Prasun Chakrabarti |
| author_sort | Aniket Vatsa |
| collection | DOAJ |
| description | Abstract The empirical application of polarization and depolarization current (PDC) measurement of transformers facilitates the extraction of critical insulation-sensitive parameters. This technique, rooted in time-domain dielectric response analysis, forms the bedrock for parameterization and insulation modeling. However, the inherently time-consuming nature of polarization current measurements renders them susceptible to data corruption. This article explores deep-learning-based short-duration techniques for forecasting polarization current to address this limitation. By incorporating spatial shortcuts, the residual long short-term memory (LSTM) network facilitates the seamless propagation of spatial and temporal gradients. Furthermore, the relative forecasting assessment of the proposed residual LSTM model’s performance is made against traditional LSTM, attention LSTM, gated recurrent units (GRU), and convolutional neural network (CNN) models. Thus, optimal model selection strategies are evaluated based on their capability to capture extended dependencies and short-term information present in the data. In addition, the Monte Carlo dropout prediction is employed to estimate uncertainty in polarization current forecasts. The findings demonstrate that the proposed residual LSTM network model for polarization current forecasting yields the lowest error metrics and maintains prediction consistency over the testing duration. Thus, the proposed approach significantly reduces PDC measurement time, providing an effective means to develop proactive maintenance strategies for evaluating the insulation condition of transformers. |
| first_indexed | 2024-03-08T12:38:55Z |
| format | Article |
| id | doaj.art-f389f519c4aa416698bf476d7c9a8465 |
| institution | Directory Open Access Journal |
| issn | 2045-2322 |
| language | English |
| last_indexed | 2024-03-08T12:38:55Z |
| publishDate | 2024-01-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj.art-f389f519c4aa416698bf476d7c9a84652024-01-21T12:17:04ZengNature PortfolioScientific Reports2045-23222024-01-0114111110.1038/s41598-023-50641-zResidual LSTM-based short duration forecasting of polarization current for effective assessment of transformers insulationAniket Vatsa0Ananda Shankar Hati1Prashant Kumar2Martin Margala3Prasun Chakrabarti4Department of Electrical Engineering, Indian Institute of Technology (Indian School of Mines)Department of Electrical Engineering, Indian Institute of Technology (Indian School of Mines)Department of Mechanical, Robotics, and Energy Engineering, Dongguk University-SeoulSchool of Computing and lnformatics, University of Louisiana at LafayetteDepartment of Computer Science and Engineering, Sir Padampat Singhania UniversityAbstract The empirical application of polarization and depolarization current (PDC) measurement of transformers facilitates the extraction of critical insulation-sensitive parameters. This technique, rooted in time-domain dielectric response analysis, forms the bedrock for parameterization and insulation modeling. However, the inherently time-consuming nature of polarization current measurements renders them susceptible to data corruption. This article explores deep-learning-based short-duration techniques for forecasting polarization current to address this limitation. By incorporating spatial shortcuts, the residual long short-term memory (LSTM) network facilitates the seamless propagation of spatial and temporal gradients. Furthermore, the relative forecasting assessment of the proposed residual LSTM model’s performance is made against traditional LSTM, attention LSTM, gated recurrent units (GRU), and convolutional neural network (CNN) models. Thus, optimal model selection strategies are evaluated based on their capability to capture extended dependencies and short-term information present in the data. In addition, the Monte Carlo dropout prediction is employed to estimate uncertainty in polarization current forecasts. The findings demonstrate that the proposed residual LSTM network model for polarization current forecasting yields the lowest error metrics and maintains prediction consistency over the testing duration. Thus, the proposed approach significantly reduces PDC measurement time, providing an effective means to develop proactive maintenance strategies for evaluating the insulation condition of transformers.https://doi.org/10.1038/s41598-023-50641-z |
| spellingShingle | Aniket Vatsa Ananda Shankar Hati Prashant Kumar Martin Margala Prasun Chakrabarti Residual LSTM-based short duration forecasting of polarization current for effective assessment of transformers insulation Scientific Reports |
| title | Residual LSTM-based short duration forecasting of polarization current for effective assessment of transformers insulation |
| title_full | Residual LSTM-based short duration forecasting of polarization current for effective assessment of transformers insulation |
| title_fullStr | Residual LSTM-based short duration forecasting of polarization current for effective assessment of transformers insulation |
| title_full_unstemmed | Residual LSTM-based short duration forecasting of polarization current for effective assessment of transformers insulation |
| title_short | Residual LSTM-based short duration forecasting of polarization current for effective assessment of transformers insulation |
| title_sort | residual lstm based short duration forecasting of polarization current for effective assessment of transformers insulation |
| url | https://doi.org/10.1038/s41598-023-50641-z |
| work_keys_str_mv | AT aniketvatsa residuallstmbasedshortdurationforecastingofpolarizationcurrentforeffectiveassessmentoftransformersinsulation AT anandashankarhati residuallstmbasedshortdurationforecastingofpolarizationcurrentforeffectiveassessmentoftransformersinsulation AT prashantkumar residuallstmbasedshortdurationforecastingofpolarizationcurrentforeffectiveassessmentoftransformersinsulation AT martinmargala residuallstmbasedshortdurationforecastingofpolarizationcurrentforeffectiveassessmentoftransformersinsulation AT prasunchakrabarti residuallstmbasedshortdurationforecastingofpolarizationcurrentforeffectiveassessmentoftransformersinsulation |